OBJECTIVES: Nivolumab has been approved globally for advanced NSCLC patients with progression on or after platinum-based chemotherapy and naïve to immunotherapy based on an overall survival (OS) benefit versus docetaxel in CheckMate 057, CheckMate 017, and CheckMate 078. This study aimed to inform evaluations of nivolumab versus treatments other than docetaxel via a network meta-analysis (NMA).

METHODS: A systematic literature review identified 77 RCTs involving treatments either recommended for chemotherapy-pretreated NSCLC or in real-world use. Eight studies were included in the analysis. Fixed effect Bayesian network meta-analysis models were used to evaluate the relative efficacy of nivolumab versus other agents as described by OS and other endpoints. Estimated hazard ratios (HRs) less than one indicate benefit with nivolumab, while HRs greater than one indicate benefit with the comparator.

RESULTS: Results from the overall population (incorporating patients of all histologies and programmed death ligand 1 [PD-L1] expression levels included in each study) suggest nivolumab improved survival versus erlotinib (in EGFR-wild type patients; Hazard Ratio [HR]: 0.51 [95% credible interval (CrI): 0.36-0.71]), ramucirumab+docetaxel (HR: 0.81 [95% CrI: 0.67-0.97]), and docetaxel (HR: 0.70 [95% CrI: 0.61-0.79]). Estimated HRs indicated similar OS efficacy of nivolumab versus both atezolizumab (HR: 0.98 [95% CrI: 0.79-1.21]) and pembrolizumab 2 mg/kg (HR: 0.98: [95%CrI: 0.77-1.25]). When limiting the analysis to the PD-L1 expressing (PD-L1 ≥1%) population (given all patients in KEYNOTE-010 were PD-L1 expressing), the estimated OS HR for nivolumab versus pembrolizumab 2mg/kg was 0.91 (95% CrI: 0.67-1.22) and the estimated OS HR for nivolumab versus atezolizumab was 0.87 (95% CrI: 0.65-1.15).

CONCLUSIONS: NMA results suggest that nivolumab improved survival versus several relevant treatments for pretreated and immunotherapy-naïve advanced NSCLC and was similar to pembrolizumab and atezolizumab in comparable populations.